Re: Constructive linguistics

Hello!

I don't see the artlang becoming the subject of an academic discipline any
time soon either.

As others have pointed out, conlangs are probably irrelevant to academic
linguistics.   FWIU (from what I understand), they aren't useful in testing
the Whorf-Sapir hypothesis.  FWIU, despite the fact that some conlangs
violate alleged linguistic universals, conlangs have yet to confront
linguists with phenomena that are as interesting as a) grammatical
structures found in natural languages or b) hypothetical nonce-grammars that
could be generated using sophisticated mathematics/computer models.

As for the humanities, I think that artngs stand at odds with the aims of
traditional literature, which generally aspires to express the most
individual outlooks in ways that make them interesting to a larger
population.   In the process of conlanging, facts that are already
interesting to a large population (everyone from professional linguists to
weekend language-trivia buffs) are used in the service of an almost purely
private endeavor.

Witness our own list.   We share information that, in one way or another, is
helpful to the individual conlanger.  We share information about natlangs,
linguistic theory, whether or not something has already been done in a
conlang, and solutions to problems that arise in conlanging.
We help conlangers do their thing.

What we don't do is acquire significant expertise in each other's conlangs.
There's nothing wrong with that.  I'd rather have some help in building my
own conlang than spend two years learning to speak and read and write in
someone else's.  So would a lot of other people.

All I'm saying is that it's hard to imagine an academic study of conlanging
as literature when even the practitioners of conlanging know only their own
conlangs and don't seem to need  opinions about the tendencies, virtues, and
values embodied across conlangs.

In fairness to artlangers who do have literary aspirations, artlang poetry
and prose has, IMO, more to recommended it than so-called "language poetry"
which, AFAIK, isn't written in any language.  More than this, perhaps the
*act* of inventing an artlang could be seen as a kind of performance art;
an artistic assertion of individuality, a refusal to be swallowed up by the
world at large.

Personally, I've always thought of conlanging as a craft, like scrimshaw.
Or model railroad building, as Jeff Henning suggested?

I like the model railroad analogy to a point--but it has one defect.  A
model railroad mimics, as much as possible, that thing that it's a model of.
The topography can be purely imaginary, and the route unnaturally short, but
otherwise, the more realistic the model train set, the more esteemed it is.

There isn't any tradition that constrains conlanging in this way.

Hence our discussion about putting tenses on nouns instead of verbs.  Hence
the fact that a conlang is not considered less worthy if it lacks the kind
of irregular forms typically found in natural languages.  Hence attempts to
design conlangs with more cases or conjugations than one finds in any
natural language.

Yes, with enough knowledge of real languages, we can strive to write a
conlang reference grammar that could pass for the reference grammar of an
obscure dialect or the recently discovered descendent of some dead language.
We can do that, but we don't have to.

Yes, with enough knowledge of the current linguistics literature, we can
strive to create a unique language whose reference grammar could,
nonetheless, pass as the reference grammar for a natural human language.
We can do that, but we don't have to.

If model railroad building were more like conlanging, the signs could be
replaced with scary idols, the station could look like a psychedelic
seashell, the ties on the tracks could be colorful little things with runes
painted on them, and the trains could be wheeled ovoids and/or polyhedrons
adorned with feathers, fur, scales, or miniature roofing shingles, along
with the occasional eyes and antennae.   Or the set could look more
realistic, but that would be up to the model builder.

Jim Grossmann

Replies